Description
When running open fga sdk with default retries configuration on production I'm occasionally getting FGA Error: socket hang up Error: socket hang up, looks like it's an error from axios library when request is timed out. Should it be added as valid case for checkIfRetryableError?
Version of SDK
v0.9.1
Version of OpenFGA (if known)
v1.11.1
Sample Code the Produces Issues
just setup client and use it with higher volume traffic (like 1k per hour)
const client = new OpenFgaClient({
apiUrl: opts.authorization.api.apiUrl,
storeId: opts.authorization.api.storeId,
authorizationModelId: opts.authorization.api.modelId,
});
Backtrace (if applicable)
Internal Server Error: FGA Error: socket hang up Error: socket hang up
at AxiosError.from (/home/app/node_modules/axios/dist/node/axios.cjs:914:14)
at RedirectableRequest.handleRequestError (/home/app/node_modules/axios/dist/node/axios.cjs:3515:25)
at RedirectableRequest.emit (node:events:536:35)
at eventHandlers.<computed> (/home/app/node_modules/follow-redirects/index.js:49:24)
at ClientRequest.emit (node:events:524:28)
at emitErrorEvent (node:_http_client:101:11)
at TLSSocket.socketOnEnd (node:_http_client:528:5)
at TLSSocket.emit (node:events:536:35)
at endReadableNT (node:internal/streams/readable:1698:12)
at process.processTicksAndRejections (node:internal/process/task_queues:82:21)
at Axios.request (/home/app/node_modules/axios/dist/node/axios.cjs:4731:41)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async attemptHttpRequest (/home/app/node_modules/@openfga/sdk/dist/common.js:188:30)
at async /home/app/node_modules/@openfga/sdk/dist/common.js:226:33
at async /home/app/node_modules/@openfga/sdk/dist/client.js:500:30
at async /home/app/node_modules/tiny-async-pool/lib/es9.js:12:34
at async consume (/home/app/node_modules/tiny-async-pool/lib/es9.js:4:30)
at async asyncPool (/home/app/node_modules/tiny-async-pool/lib/es9.js:21:11)
at async OpenFgaClient.batchCheck (/home/app/node_modules/@openfga/sdk/dist/client.js:504:26)
Expected behavior
Error should be retired
Description
When running open fga sdk with default retries configuration on production I'm occasionally getting
FGA Error: socket hang up Error: socket hang up, looks like it's an error from axios library when request is timed out. Should it be added as valid case for checkIfRetryableError?Version of SDK
v0.9.1
Version of OpenFGA (if known)
v1.11.1
Sample Code the Produces Issues
just setup client and use it with higher volume traffic (like 1k per hour)
Backtrace (if applicable)
Expected behavior
Error should be retired